WAGENINGEN UNIVERSITEIT/ WAGENINGEN UNIVERSITY LABORATORIUM VOOR ENTOMOLOGIE/ LABORATORY OF ENTOMOLOGY Testing the ‘usurpation hypothesis’ with the secondary hyperparasitoid Lysibia nana No.: 07.10 Name: Martine Kos Period: November 2006 – April 2007 1st Examinators: Jeff Harvey and Martijn Bezemer 2nd Examinator: Marcel Dicke Content Abstract..................................................................................................................................... 2 Introduction .............................................................................................................................. 2 Hypotheses ................................................................................................................................6 Material and methods .............................................................................................................. 8 Insect species.......................................................................................................................... 8 Insect cultures......................................................................................................................... 9 Experiments............................................................................................................................ 9 Statistical analysis ................................................................................................................ 14 Results ..................................................................................................................................... 15 Experiment 1: Parasitoid survival ........................................................................................ 15 Experiment 2: L. nana choice experiment ........................................................................... 23 Experiment 3: L. nana behaviour observation ..................................................................... 25 Experiment 4: Caterpillar survival ....................................................................................... 29 Discussion................................................................................................................................ 30 Outlook for future research .................................................................................................. 33 Acknowledgements................................................................................................................. 33 References ............................................................................................................................... 34 1 Abstract Parasitoids regulate the host during their development to increase parasitoid survival and fitness. Larvae of the primary parasitoid Cotesia glomerata feed primarily on the haemolyph and fat body of the host during their development. C. glomerata has evolved several mechanisms to keep its host (larvae in the genus Pieris, with its main host P. brassicae) in a suitable condition during the larval development. After egression, the larvae spin their cocoons close to the host. However, due to selective tissue destruction, the host does not die after parasitoid emergence, but remains close to the cocoons and sometimes even coils over the cocoons. The host caterpillar spins a silk layer to cover the cocoons, and might react aggressively when disturbed. The ‘usurpation hypothesis’ theory suggests that C. glomerata might be able to manipulate and use the behaviour of the surviving host after egression and pupation as a defence against predators and secondary hyperparasitoids. In this study, the ‘usurpation hypothesis’ has been tested with the secondary hyperparasitoid Lysibia nana. Different aspects of the host, being the presence, state and position of an attending caterpillar and the presence of a silk web covering the cocoons, were tested separately. The ‘usurpation hypothesis’ was tested on a number of different levels, being parasitoid survival, choices made by L. nana and L. nana foraging behaviour. As hypothesised, the presence, state and position of a caterpillar had no influence on parasitoid survival, or on L. nana foraging behaviour. Opposite to expectations, the presence of a silk web increased C. glomerata survival, and decreased L. nana survival. Furthermore, L. nana showed less oviposition behaviour on the silk-covered cocoons. However, L. nana did not prefer to stay on the bare cocoons compared to the covered cocoons. Potential olfactory cues emitted by the host or host by-products did not attract L. nana. The results of this study provide partial support for the ‘usurpation hypothesis’ in this multitrophic association: the presence of a silk web increased parasitoid survival, but the caterpillar itself did not increase parasitoid survival and not all caterpillars spin a silk web. Therefore the survival and presence of the host after parasitoid emergence did not necessarily lead to a higher parasitoid survival. Furthermore, since hyperparasitoids are much specialised and abandoning a potential host could decrease reproductive success significantly, which was also supported by the results, it is expected that the usurpation of host behaviour by parasitoids is much more aimed at protection from predators, instead of hyperparasitoids. Introduction Parasitic wasps lay their eggs inside or externally on a host and during their larval development the parasitoids feed on the haemolymph and other internal tissues of the host (Godfray 1994; Brodeur and Vet 1994; Harvey 2000). Afterwards, the larvae pupate and new parasitoids emerge from the cocoons (Godfray 1994; Harvey 2000). Parasitoids regulate the host during their development to increase parasitoid survival and fitness, and they influence the development, behaviour, physiology and morphology of the host (Vinson and Iwantsch 1980; Slansky 1986). These effects on the host are often due to factors such as polydnaviruses and virus-like particles injected into the host by the ovipositing female (Beckage and Gelman 2004), but can also be due to other parasitoid-derived products such as teratocytes (Dahlman 1991; Beckage and Gelman 2004). Most species of parasitic wasps normally attack only a few host species (Godfray 1994; Harvey and Witjes 2005). They are much more specialised than arthropod predators, most of which are generalists and thus attack many different kinds of prey (Godfray 1994; 2 Harvey and Witjes 2005). Parasitoids exploit a highly nutritious resource, the host body. Furthermore, the costs of metabolic activity are small compared to actively foraging predators, because the parasitoid larvae are sessile (Slansky 1986). However, parasitoids depend on the resources provided by a single host, whereas predators need many prey items to reach maturity. A parasitoid must therefore make sure to optimally utilise this limited resource (Harvey et al. 2004b). A primary parasitoid usually parasitizes the larvae of an herbivorous insect, whereas a hyperparasitoid parasitizes a primary parasitoid. Within the hyperparasitoids, a primary hyperparasitoid parasitizes the parasitoid larvae that are still inside their host and a secondary- or pseudohyperparasitoid parasitizes newly cocooned pre-pupae and pupae of a primary parasitoid (Harvey et al. 2003). At the end of their larval development the mature parasitoid larvae egress from the host and start to spin a cocoon to pupate within (primary parasitoids), or they pupate within the cocoon that the host had spun (primary (with a Lepidopteran secondary host) and secondary hyperparasitoids) (Harvey et al. 2003). Two broadly different macroevolutionary host usage strategies have been described among parasitic wasps. Koinobiont parasitoids (including primary parasitoids and primary hyperparasitoids) develop in hosts that continue feeding, growing, and defending themselves during the early phases of parasitism, whereas idiobiont parasitoids (including secondary hyperparasitoids) develop in non-growing host stages, such as eggs or pupae (Askew and Shaw 1986; Harvey 2005; Harvey et al. 2006). Because idiobiont hosts do not feed or grow, the host resources are effectively ‘static’ and the development of the parasitoid is dependent on the quality and amount of resources available at the time of oviposition (Mackauer and Sequeira 1993; Harvey 2005). Within idiobiont parasitoids, larger hosts are assumed to be of higher quality than small hosts, because they contain more resources for parasitoid development (Harvey et al. 2004b; Harvey 2005). However, as hosts age, they undergo dramatic morphological and physiological changes (sclerotization of the cuticle and differentiation of body parts into recognizable structures such as wings and antennae) and this inhibits the rate of consumption and digestion by the parasitoid larva. Therefore, older hosts will be less suitable for parasitoid development than younger hosts, even when these younger hosts are smaller (Harvey 2005; Harvey et al. 2006). Koinobiont hosts continue feeding and growing and the relationship between host quality and parasitoid fitness is much more complicated than is the case with idiobionts (Mackauer and Sequeira 1993; Harvey 2005). Amongst koinobionts, the size of emerging adults often increases with host size at parasitism, but mortality in larger hosts can also be also higher (Harvey et al. 2004b). Consequently, overall parasitoid fitness may be a dome- shaped function of initial host size, which means that parasitoid fitness is optimized in hosts of intermediate size at parasitism